Technical Support
Contact Us
Anergy Firsts
•    Indigenous temperature and
     humidity controllers
•    Suction Guide
•    Test Points
•    Data Logger
•    Reflex
•    Hydronic Pressurisation System
•    Indigenous BMS
•    Thermal Energy Meter
When will HVAC Controls become intelligent

The control technology has been changing at fast pace in the last few decades, at least that is what every one claims or believes, but you scratch the surface and you find that nothing much has changed. All the changes that have taken place are superficial and only skin deep. The people still complain as much about uncomfortable conditions as in good old days of mechanical thermostat.

The introduction of the latest, state-of-the-art, controls with space age technologies has hardly made any impact on the occupant comfort. The technology has been used to only glamorize the controls and improve the front ends but deep inside it's the same old algorithms.
Why the Controls have not changed
Now, the most important question, why the control strategies have not changed keeping up with time? For sure, this issue is not being discussed for the first time.
But many a times vested commercial interests overshadow the technical developments. Recently I came across an article in an Indian Engineering magazine written by an executive of a leading controls manufacturer. In this article several points considered to be important in selecting a BMS vendor have been listed and one of them is backward integration. Actually speaking this term backward integration has been one of the main reasons of backwardness in the field of HVAC controls.
For more than 20 years that microprocessors have been available to HVAC control manufactures, few have taken the bold initiative to develop new control strategies to improve occupant comfort and energy conservation. An overwhelming majority of control manufacturers decided to configure the DDC controls as closely as possible to the pneumatic or analog electronic controls they were replacing. This was done to assure their existing customers that new developments are not going to affect their present installations and future upgrades shall be possible.
Why the Controls should change
Take a look at the most humble HVAC control, the thermostat of an air-conditioner. Earlier it used to use a bimetallic strip to control temperature, now it is an electronic device with a sleek looking remote control and a temperature display but it still makes you feel equally uncomfortable between each ON & OFF cycle. Has technology really changed anything? Can temperature display reduce the discomfort between cycles? No prizes for guessing the correct answer.
Some people may say that thermostat's sole purpose is to keep the room temperature constant and the present generation of controls certainly does it better, but is there not a major error in thinking. The actual objective of a thermostat is not to keep temperature constant but to maintain maximum occupant comfort. Room temperature does not always correspond to the subjective temperature felt by people. The people's perception is affected by the air circulation, thermal radiation and the brightness, as well.
HVAC systems are highly dynamic and have non-linear characteristics. The loads, internal as well as external, keep changing with day times and seasons. PID control algorithms, in use even today, were developed in the era of pneumatic or analog electronic controls, when it was almost impossible to implement a more advanced algorithm. These control algorithms assume linear relationship between the controlled variable and the output, which is rarely true for HVAC application.
Take the case of modulating temperature control system, it has a blower, coil, valve, controller, and a source of hot or cold water. The control system is simplified by assuming constant air volume, constant water temperature to the coil and constant water entering and leaving pressures. This simplification allows to add some heat if it is too cold or to take away some heat if it is too hot, by changing the flow of water to coil. The controller relies on a simple proportional logic, the controller output changes in direct proportion to the error between the set point and the measured value. It is a linear function and assumes that 20% output will produce 20% heat transfer whereas the actual heat exchange process tends to be non-linear. Typical coil characteristics shown in fig.1 show that 20% flow to coil yields far greater heat transfer, almost 60%.
Both NO & NC contacts are SPST type, the difference being that when the external force is removed the NO type of contact switches to open position whereas the NC type of contact switches to closed position.
The solution relied on is to use a valve plug with equal percentage characteristics so that the combined system (valve plus coil) has an overall linear characteristics. In practice, the coil characteristics changes with airflow and water supply temperature, the valve characteristics changes with pressure and it is never possible to perfectly match the valve with the coil. As a result, the characteristics of the combined system changes a lot more under varying load conditions. In other words you can never set the system to achieve uniform, compatible set of conditions and once again have to resort to resetting the set points often. BMS may come handy if you have one and save you the trouble of going to each AHU to reset the set point, like the friendly air-conditioner remote.
Recently a very senior HVAC consultant lamented that even the most sophisticated BMS systems today are barely able to achieve any more functionality than a CLIP (Centralized Light Indication Panel) he designed more than 30 years back. The only difference being that the customer pays much more, mainly for the high-end graphics.
How the Controls should change
In future the users will become more discerning and demand more comfortable conditions, and to achieve these controls will have to think like humans. Human comfort is all about feelings, numeric values have little relevance. HVAC controls of future will have to understand the feelings of the occupant, not the other way round, where the occupant has to feed in a data, like temperature, at which he feels he may be comfortable.
The HVAC Controls will use microprocessors operating at high speeds with large memories and tremendous computing capabilities. They will be capable of storing & analyzing the linear as well as non-linear process data.
The parameters influencing human comfort are many, like
- Temperature
- Type of clothing
- Air circulation
- Radiation
- Brightness
- Time of the day
- Duration of exposure
- Activity level
- Humidity
Measuring all these parameters accurately and then generating an algorithm to calculate output will not only make the system prohibitively expensive but will also tax the operating efficiency of the fastest microprocessor. To avoid this controls will have to become intuitive and learn from experience. The controls will need to imitate the human logic. Have you ever wondered how fast and easily you are able to control mixer tap water temperature and that too without measuring the actual temperature. This logic is based on approximate reasoning and is called FUZZY LOGIC.
Fuzzy logic provides a simple way to arrive at a definite conclusion based on vague, ambiguous, imprecise, noisy or missing input information. Instead of well-defined values it uses vague input definitions such as VERY HOT, HOT, NORMAL, COOL, VERY COOL. Similar FUZZINESS is found in almost all our decisions, in our thinking, in the way we process information. Fuzzy Logic incorporates a simple, rule based IF X AND Y THEN Z approach to solving a control problem rather than attempting to model a system mathematically.
Examples of fuzzy rules would be:
- If the temperature is quite a bit above the set point, then increase cooling
- If the temperature is a little above the set point and rising at a medium rate, then increase cooling slightly.
- If the temperature is somewhat above the set point, but constant, then wait a while to change cooling.
More complex fuzzy rules would be
- If the temperature is little above the set point and humidity is below normal and outside temperature is below normal and the temp is constant, then reduce the cooling a little.
- If the temperature is above the set point and is rising at slow rate and the last increase in output to the control valve did not affect the rise rate much, then increase the output by a value more than the last increment.
Fuzzy logic relying on approximate reasoning uses numerical equivalents of terms like long, short, slow, quick, small, large, open, close to determine the correct quantum and mode of output. Fuzzy logic is able to accept variables such as deviation from set point, direction & rate of change, historical average output, the time elapsed since the last change, effect of last change in output etc.
Properly tuned fuzzy logic controllers are superior to conventional PID control algorithms, because they respond more quickly to a set point and environmental change and have minimal overshoot of controlled variable. The controlled systems reach a steady state faster with reduced oscillations about set point. The quick response improves comfort and, since the system spends less time in transient operation, energy consumption is reduced. Adaptive fuzzy logic controllers automatically improve their performance until they converge to a predetermined optimal condition.
Fuzzy logic is not very new, many of fuzzy logic based commercial products like, self-focusing cameras, washing machines, automobile engine controls, anti-lock braking systems, are already available and are successful. HVAC controls manufacturers have been very slow to adopt, however, they will soon be forced to change their attitude by the rising customer expectations.
© Copyright 2010 Anergy Instruments Pvt. Ltd. All right's reserved.